Chester Ismay (Portland State University), Arturo Valdivia (Indiana University)
Abstract
This workshop offers a roadmap for designing and teaching a hybrid introductory course that integrates foundational statistics with modern data science. Drawing on the pedagogical strategies in ModernDive (https://moderndive.com/v2), participants will explore how core topics—such as descriptive analytics, linear regression, and simulation-based methods for confidence intervals and hypothesis testing—fit into a broader data science framework. By combining these statistical concepts with hands-on coding in R and the tidyverse, instructors can help students develop both statistical reasoning skills and computational fluency from the start.
Through concrete examples using resampling methods including shuffle-and-compare and coding walkthrough activities, we illustrate bootstrap and permutation tests and demystify simulation-based inference while simultaneously building students’ confidence in data manipulation, visualization, and modeling. Participants will learn an incremental approach to expanding core ideas throughout the course, laying the groundwork for more advanced analytical techniques.
By the end of the workshop, attendees will have a clear plan for delivering a cohesive, skills-focused course at the intersection of statistics and data science. This integrated approach ensures that students—and their instructors—gain adequate statistical foundations and coding proficiency while continually progressing toward the data-driven and reproducible practices needed for success. Empowering students with programming skills for efficient problem-solving, creativity, and a strong foundation in statistical reasoning is essential across all sectors.